After decades of stability from the 1920s to the early 1970s, the rate of incarceration in the United States has increased to a rate more than four times higher than in 1972. In 1972, the U.S. incarceration rate—the number in prisons and local jails per 100,000 population—stood at 161. After peaking in 2009, the number of people in state and federal prisons fell slightly through 2012. Still, in 2012, the incarceration rate was 707 per 100,000, a total of 2.23 million people in custody (Glaze and Herberman, 2013). With nearly 1 of every 100 adults in prison or jail, the U.S. rate of incarceration is 5 to 10 times higher than the rates in Western European and other liberal democracies.1
The large racial disparity in incarceration is striking. Of those behind bars in 2011, about 60 percent were minorities (858,000 blacks and 464,000 Hispanics) (Carson and Sabol, 2012; Maguire, n.d., Table 6.17.2011). The largest impact of the prison buildup has been on poor, minority men. African American men born since the late 1960s are more likely to have served time in prison than to have completed college with a 4-year degree (Pettit and Western, 2004; Pettit, 2012). And African American men under age 35 who failed to finish high school are now more likely to be behind bars than employed in the labor market. The rise in incarceration
1The 1972 incarceration rate is calculated from counts of the prison and jail population reported in Hindelang et al. (1977, Tables 6.1 and 6.43). The 2012 prison and jail incarceration rates and the incarcerated population are reported in Glaze and Herberman (2013, Table 2). International incarceration rates for European countries are from Aebi and Delgrande (2013), and data on Australia, Canada, New Zealand, and the United States are from Walmsley (2012), as well as updates from the International Centre for Prison Studies (2013).
transformed not only the criminal justice system, but also U.S. race relations and the institutional landscape of urban poverty.
The U.S. justice system has charted a unique path to crime control that traverses poor and minority communities across the country. A number of research literatures explore the sources of rising incarceration rates in policy and social change; the relationship between crime and incarceration; and the effects of incarceration on the employment, health, and family life of the formerly incarcerated and more broadly on U.S. civic life. Yet there has been no comprehensive effort to date to assess the causes, scope, and consequences of contemporary incarceration rates. Four questions stand out:
1. What changes in U.S. society and public policy drove the rise in incarceration?
2. What consequences have these changes had for crime rates?
3. What effects does incarceration have on those in confinement; on their families and children; on the neighborhoods and communities from which they come and to which they return; and on the economy, politics, structure, and culture of U.S. society?
4. What are the implications for public policy of the evidence on causes and effects of high levels of incarceration?
An ad hoc committee of the National Research Council was asked to review the scientific evidence on these questions. By weighing the evidence on both the causes and consequences of high rates of incarceration, the committee’s work may help the public and policy makers decide whether the current rates are too high and, if so, to explore policy alternatives. And if the evidence is wanting or inconsistent, then this study can indicate directions for future research.
The committee’s statement of task (see Box 1-1) describes in greater detail the scope of its efforts.
Incarceration is a unique state function. The forcible deprivation of liberty and detention in a facility designed for the purpose is a restriction on the individual freedom to which liberal societies aspire. Incarceration represents a collective decision that some among us are too dangerous, or their crimes too serious, to circulate freely in the community. To preserve order and safety, to affirm norms of lawful conduct, and to help remedy criminal behavior, we built lockups, detention centers, asylums, jails, and prisons. These institutions reflect how a society through its political process has negotiated a compromise between order and freedom. Incarceration is in many ways a foundational institution, being the last resort of a state’s
authority in the performance of its many other functions. As many have remarked, the use and character of incarceration thus reveals something fundamental about a society’s level of civilization and the quality of citizenship (de Beaumont and de Tocqueville, 1970; Churchill, 1910; Dostoevsky, 1861).
Imprisonment lies at one end of a continuum of legally ordered restraints on liberty, some imposed not as punishment but as part of the criminal justice process prior to a possible charge or conviction or following release from confinement. At the other end of the continuum are forms of community supervision, such as probation and parole, as well as in-home detention. Compared with punishment outside a facility, confinement in a local jail or detention facility for short periods is likely to be seen and experienced as a more severe form of punishment. Prison incarceration then follows. Of course, imprisonment itself varies in severity, depending on the specifics of confinement and treatment.
Contemporary incarceration takes many forms. The juvenile justice system has developed a special set of rules and protocols for the detention of children. Criminally convicted adults are held in state or federal prisons, generally serving more than a year for a felony. Local jails typically detain those serving short sentences or awaiting trial. Immigrants awaiting deportation may be held in federal detention facilities. The mentally ill may be housed under civil commitment to state hospitals. The conditions of incarceration, like those who are confined, thus show considerable variation.
The primary focus of this report is on adults incarcerated in prisons and jails. Prisoners form an important subset of this group because of their large numbers among the total incarcerated population and the long terms of their confinement. Throughout the report, therefore, we often focus specifically on prison incarceration. The size of the prison population depends not only on the number of crimes, arrests, convictions, and prison commitments but also on the use of alternatives to incarceration and on responses to violations of the terms of parole or probation. Although other forms of incarceration outside of the adult criminal justice system are undeniably important, they have been less important to the steep rise in incarceration over the past four decades that forms our main charge.
This study differs from many conducted by committees of the National Research Council with respect to its scope and the number of questions posed about a complex social phenomenon. The causes of the increase in incarceration rates are disputed, and its consequences are not fully understood. Nevertheless, a rigorous review of the evidence is now timely. A burgeoning research literature helps explain why the U.S. incarceration rate grew so dramatically and examines the consequences for those incarcerated, their families, communities, and U.S. society. For several decades, enthusiasm for incarceration dominated crime policy and the related public
Causes and Consequences of High Rates of Incarceration Statement of Task
An ad hoc panel will conduct a study and prepare a report that will focus on the scientific evidence that exists on the use of incarceration in the United States and will propose a research agenda on the use of incarceration and alternatives to incarceration for the future. The study will explore the causes of the dramatic increases in incarceration rates since the 1970s, the costs and benefits of the nation’s current sentencing and incarceration policies, and whether there is evidence that alternative policies would more effectively promote public safety and community well-being.
Recognizing that research evidence will vary in its strength and consistency, the panel will undertake the following tasks:
1. Describe and assess the existing research on the causes, drivers, and social context of incarceration in the United States over the past 30-40 years. To what extent does existing research suggest that incarceration rates were influenced by historical and contemporary changes in:
a. operations of criminal justice system and other public sector systems that may affect rates of arrest or conviction, and nature and severity of sanctions: such as patterns of policing, prosecution, sentencing, prison operations, and parole practices;
b. legal and judicial policies, such as changes in law, institutional policies and practices, and judicial rulings affecting conditions for arrest, sanctions for various crimes, drug enforcement policies, and policies regarding parole and parole revocation; and
c. social and economic structure and political conditions, such as criminal behavior, cultural shifts, changes in political attitudes and behavior, changes in public opinion, demographic changes, and changes in the structure of economic opportunity.
conversation. Commentators on both the left and the right are now reacting critically to the incarceration boom, partly out of concern for growing correctional budgets, partly because of questions about the effectiveness of incarceration in reducing crime, and partly out of misgivings about the values that have come to dominate penal policy (e.g., Gramlich, 2013; Kabler, 2013; Alexander, 2010). Reform, it appears, is under way. At the state level and in the federal government, many elected officials are supporting initiatives aimed at reducing prison populations and are turning to the research evidence for guidance. In this context, the committee hopes to inform a critical conversation about the significance of high incarceration rates for U.S. society and the future of the nation’s penal and social policies.
2. Describe and assess the existing research on the consequences of current U.S. incarceration policies. To what extent does the research suggest that incarceration rates have effects on:
a. crime rates, such as to what extent this is due to deterrence and incapacitation, to rehabilitation, or to criminogenic effects of incarceration;
b. individual behavior and outcomes, during imprisonment and afterward, such as changes in mental and physical health, prospects for future employment, civic participation, and desistance/reoffending;
c. families, such as effects on intimate partners and children, patterns of marriage and dating, and intergenerational effects;
d. communities, such as geographic concentrations, neighborhood effects, effects on specific racial and ethnic communities, high rates of reentry and return in some communities, labor markets, and patterns of crime and policing; and
e. society, such as (in addition to effects on the crime rate) the financial and economic costs of incarceration, effects on U.S. civic life and governance, and other near-term and longer-term social costs and benefits.
3. Explore the public policy implications of the analysis of causes and consequences, including evidence for the effectiveness and costs of alternative policies affecting incarceration rates. What does the research tell us about:
a. efficacy of policies that may affect incarceration or serve as alternatives to incarceration, including their effects on public safety and their other social benefits and costs;
b. the cost-effectiveness of specific programmatic approaches to reducing the rate of incarceration;
c. how best to measure and assess the potential costs and benefits of alternative policies and programs; and
d. ways to improve oversight and administration of policies, institutions, and programs affecting the rate of incarceration.
To address the study charge, the National Research Council assembled a committee of 20 scholars and practitioners to review and assess the research evidence. The committee members include not only criminologists and sociologists who have conducted original research on these issues but also representatives of other academic disciplines, including economics, political science, psychology, law, medicine, and history, who brought to bear different methods and perspectives. The members include those whose professional experience gives them practical insights into the workings of the judicial and corrections systems and the policy debates in the legislative and executive branches of government. To aid in this study, the committee also enlisted several other scholars with specialized expertise to review
particular subsets of questions, such as the health and public health implications of incarceration; the policy shifts and system dynamics leading to higher incarceration rates; and the availability in prisons of rehabilitative education, training, treatment, and work experience.
We hope the result of the committee’s work will be seen as a fair summary of what is known today about the sources of the rise of incarceration in the United States; how it has affected people, communities, and society; and the implications of that knowledge for public policies determining future rates of incarceration.
A central question for public policy is whether increasing the incarceration rates affect public safety and, conversely, whether crime rates contributed to the growth of imprisonment in America. The historical relationship between crime and incarceration is complex. On the one hand, the decades-long rise in incarceration rates began following a substantial rise in crime rates in the United States. Yet the growth of the prison population continued through and after a major decline in crime rates in the 1990s. In reviewing the evidence, the committee paid attention to the effects of changes in state and federal policy and practice over the period of the rise in incarceration rates, including the relationship of policy changes to crime rates. We noted that in many instances, these policy choices reflected and resulted from broader political and social currents, including, for a variety of reasons, a marked tendency to resort to imprisonment, and harsh punishments generally, as society’s preferred response to crime, even when crime rates were falling.
Understanding the impact of high rates of incarceration on crime is challenging. Incarceration can reduce crime by incapacitating those who would otherwise be committing crimes in free society. Incarceration may also deter or rehabilitate those who are punished from committing future crimes. Fear of such punishment may deter others from committing crimes. On the other hand, the prison experience and its aftermath may in some cases contribute to future criminal activity. The net effect of incarceration on crime will vary depending, for example, on who is sent to prison, the type of crime, the length of sentences, and how people are treated while in prison and after release. Despite a large and growing body of studies exploring the complex relationship between crime and incarceration rates over recent decades, then, a precise quantification of the impact of high rates of incarceration on U.S. crime rates remains a significant scientific challenge.
In this report, we are not simply concerned with explaining changes in the rates of incarceration. Nor are we limited to analyzing the effects of imprisonment on individuals who serve prison sentences during the era of high incarceration. We also consider the aggregate, cumulative effects of the nation’s incarceration policies. America’s high rates of
incarceration have changed the meaning and consequences of a prison sentence for those who go to prison and for the families and communities to which they return. Over time, high incarceration rates may increase or decrease public safety, alter the functioning of labor markets and the economy, strengthen or weaken the fabric of communities, and skew the distribution of income and opportunity. Higher rates of incarceration also affect U.S. civic life, influence the nation’s pursuit of racial justice, and tip the balance in close elections. At the most basic level, more incarceration uses resources that could be spent for other purposes. Finally, we also assess the evidence on how high incarceration rates and their consequences affect the quality of American democracy.
Incarceration—legally imposed deprivation of personal liberty, typically in a facility specially designed for the purpose—is one of the most severe forms of punishment a society can impose. Prison terms usually are reserved for those found guilty of more serious crimes, defined as felonies by state and federal legislatures.
The scale of incarceration can be measured in a variety of ways. The incarceration rate is usually presented as a ratio of those in prison (or prison and jail) at a given time to a society’s (or state’s) population. The incarceration rates can be calculated for specific demographic groups in the population—by race or age, for example—and for small geographic areas, such as neighborhoods or blocks. The incarceration rate describes the footprint of the penal system in society. The magnitude of incarceration also might be measured by scaling prison admissions by crimes or arrests rather than by population. Such measures reflect the impact of prosecution and sentencing policies on the overall punitiveness of the criminal justice system. Both kinds of statistics are reported in Chapter 2.2
We are concerned in this report not only with the numbers behind bars but also with the nature and meaning of that experience and how it has changed over the period of the rise in incarceration rates. How one views the increasingly frequent resort to prison in the United States also depends in part on how one understands the purposes served by imprisonment for society and for the sentenced individual.3
2The committee considered and rejected the notion that the incarceration rate might also be presented as a ratio of those in prison to crimes reported. There is no analytical connection between one year’s crimes and a prison population sentenced for crimes committed years ago.
3A convenient summary history of thinking about incarceration and its uses can be found in Simon and Sparks (2013, Chapters 1-7).
Crime and punishment are social and legal constructs. Their nature and meanings change over time and differ from one society (and one person) to another. In American jurisprudence, a prison sentence serves three possible purposes. First, the purpose of a prison sentence may be understood primarily as retribution, or “just deserts,” meaning that the severity of a given crime requires deprivation of the liberty of the person found guilty of that crime. Second, a prison sentence may be justified as a way of preventing crime, either through deterrence of the individual sentenced (specific deterrence), deterrence of others in society at large who may be inclined to offend (general deterrence), or avoidance of crimes that might otherwise have been committed by that individual absent incarceration (incapacitation). Finally, a prison sentence may be deemed justified as a means of preventing future crimes through the rehabilitation of the individual incarcerated. Of course, these rationales are not mutually exclusive.
Throughout U.S. history, the emphasis on one or another rationale for incarceration has shifted significantly, and it continues to change. As a consequence, the conditions of confinement and the experience of returning to society also have changed. To understand the effects of the rise in incarceration, one must examine how prison environments have changed as the numbers of prisoners have increased and how this changed environment may lead to different outcomes for the individuals incarcerated.
Whereas the jurisprudence of incarceration emphasizes the purposes of retribution, deterrence, incapacitation, and rehabilitation, criminal punishment also provides a vivid moral symbol, publicly condemning criminal conduct. Thus the French sociologist Emile Durkheim (1984) argued that penal law affirms basic values and helps build social solidarity. By this account, punishment activates society’s moral sentiments and reinforces the collective sense of right and wrong. Critics have objected that, rather than reflecting “society as a whole,” institutions of punishment under real conditions of social and economic inequality burden the disadvantaged (Spitzer, 1991; Lukes and Scull, 1983; Garland, 2013). From this perspective, prisons and jails reflect and perhaps exacerbate social inequalities rather than promote social solidarity. Legal principle has grappled with the penal system’s innate potential for injustice. Rules of constraint were developed to restrict the unbridled and arbitrary application of punishment. Expressed in the language of Western jurisprudence, justice requires that society’s decision to deprive a citizen of liberty through imprisonment be constrained by two countervailing principles: proportionality (punishment should be tailored to the severity of the crime) and parsimony (punishment should not be more severe than required to achieve a legitimate public purpose) (see the discussion of guiding principles below).
Some scholars have argued that, in light of this nation’s long history of troubled race relations, it is especially important to consider whether
prison and other punishments unfairly burden African Americans and other minority groups. If so, the justice system only reinforces historical inequalities, thereby undermining the social compact that should undergird the nation’s laws. Other scholars have stressed the utilitarian value of prison for achieving socially desired ends. From this societal viewpoint, the use of incarceration is assessed according to whether its social benefits exceed its social costs. By this instrumental view, imprisonment can be used, for example, to contain and discourage crime—directly by confining those prone to commit further crimes or by deterring them or, by example, others from committing future crimes. Assessments of the effectiveness of policies favoring incarceration would therefore depend on an empirical understanding of its purported benefit of crime prevention or other social benefits, weighed against the direct costs of the prisons themselves and the indirect social costs incurred by removing incarcerated individuals from society.
Prisons also can support the rehabilitation of those incarcerated so that after release, they are more likely to live in a law-abiding way and reintegrate successfully into the rhythms of work, family, and civic engagement. In this narrower view of the instrumental value of incarceration policies, the effectiveness of prisons is measured by such outcomes as lower rates of recidivism and higher rates of employment, supportive family connections, improved health outcomes, and the standing of the formerly incarcerated as citizens in the community. The relevant scholarly literature focuses on issues of the availability and effectiveness of programs; the impact of the prison environment on the self-concept, behavior, and human capital of those incarcerated; and the experience of leaving prison and returning home.
Yet another stream of scholarly inquiry examines the role of the criminal justice system, and in particular the role of prisons, in controlling entire categories or communities of people. In this view, the laws of society and the instruments of punishment have been used throughout history to sustain those in power by suppressing active opposition to entrenched interests and deterring challenges to the status quo. This scholarly literature has examined the role of the justice system—including the definition of crimes by legislatures, enforcement of laws by the police, and uses of incarceration—in dealing with new immigrant groups, the labor and civil rights movements, the behavior of the mentally ill, and the use of alcohol and illegal drugs, to cite some examples. In recent years, scholars in this tradition have focused on the impact of the justice system on racial minorities in the United States and specifically on the impact of recent high rates of incarceration on the aspiration for racial equality. Researchers who study the power relations of society reflected in the criminal justice system often observe that the poor, minorities, and the marginal are seen as dangerous or undeserving. In these cases, the majority will support harsh punishments entailing long sentences and the use of imprisonment for lesser offenses. The effect of incarceration
and other punishments used in this way may be to reproduce and deepen existing social and economic inequalities.
Because incarceration imposes pain and loss on both those sentenced and, frequently, their families and others, these costs also must be weighed against its social benefits in determining when or whether the deprivation of liberty is justified. This equation must consider as well the harms caused by those sent to prison. These harms include those experienced by individual victims and the broader negative social effects emanating from the criminal act. By punishing breaches of the social compact, vindicating the victims of crime often is viewed as an important purpose of the criminal sanction. Such an analysis of costs and benefits must be both normative and empirical. While there are no scientific solutions to normative problems, evidence on the effects of incarceration can inform that analysis. Moreover, given the pain imposed by imprisonment and other harsh punishments, it may be reasonable to minimize their use when alternatives can achieve the same social benefits at lower cost to society. High incarceration rates may signal that in many instances, prison is being used when alternatives would achieve equal or better outcomes for society.
Because incarceration encompasses a range of experiences that vary widely across individuals and from one era or place to another, its effects are difficult to assess. This variation arises not only from differences in the legal terms of sentences, such as length and conditions for release, but also from differences in the conditions of confinement and after release. Harsh or abusive prison environments can cause damage to those subjected to them, just as environments that offer treatment and opportunities to learn and work can provide them with hope, skills, and other assets. So while we talk about incarceration as a single phenomenon, it in fact describes a wide range of experiences that may have very different effects.
For each set of questions posed in its statement of task, the committee reviewed and weighed the published research and, where the evidence permitted, summarized what is known about the phenomenon of high rates of incarceration, its causes, its effects, and the implications of that knowledge for public policy. In many respects, the body of published research on these topics is now substantial and continues to grow quickly. On some questions, the weight of evidence from empirical studies is compelling. For others, it is suggestive but not definitive. In still other cases, it is thin or conflicting. An important part of our work involved identifying the limits of current knowledge and therefore of its usefulness as a guide to the public and policy makers.
In light of the challenges to empirical research in this area, our
approach was first to identify the strongest, most methodologically rigorous individual studies and then, where possible, to review multiple studies using differing methods to establish the extent to which there is compelling evidence, and some degree of agreement, on the causes or effects of high rates of incarceration in the United States. Although there is something to be learned from the experience of other affluent societies that have followed a different path, our main point of reference was earlier in this country’s history, when incarceration rates were a fraction of what they are today.
A discussion of values has been notably missing from the nation’s recent policy debates on the use of prison. Although policies on criminal punishment necessarily embody ideas about justice, fairness, and desert, the recent policy discourse often has been characterized by overheated rhetoric or cost-benefit calculations that mask strong but hidden normative assumptions. Basic principles for penal reform should be transparent and open to debate.
In the period of rising incarceration rates and public concerns about safety, elected officials and other policy makers have argued that those committing crimes should be held accountable and punished severely. These values of offender accountability and crime control have become paramount, and older principles that balance the tendency to harsh punishment have receded from the policy debate. In undertaking this study, the committee reviewed the scholarly literature on the role of prison in society and the principles governing correctional policy generally. Based on this review, the committee articulated a set of guiding normative principles that, if observed, would restore balance to the discussion of criminal justice values. The following four normative principles helped the committee interpret the scientific evidence and guided the committee in carrying out its charge to recommend new policy alternatives:
1. Proportionality: Criminal offenses should be sentenced in proportion to their seriousness.
2. Parsimony: The period of confinement should be sufficient but not greater than necessary to achieve the goals of sentencing policy.
3. Citizenship: The conditions and consequences of imprisonment should not be so severe or lasting as to violate one’s fundamental status as a member of society.
4. Social justice: Prisons should be instruments of justice, and as such their collective effect should be to promote and not undermine society’s aspirations for a fair distribution of rights, resources, and opportunities.
Rather than describing the positive goals of imprisonment, each of these principles describes a different kind of constraint. These principles also set firm limits on one of the main impulses behind the dramatic rise in incarceration—the desire for retribution. We recognize that the urge to express public disapproval of criminal behavior is a legitimate purpose of punishment, but the disapproval of crime must be expressed within the bounds set by other normative convictions. The state’s authority to deliberately deprive people of their liberty through incarceration may be abused, and its misuse may undermine its legitimacy. We elaborate on the scholarly basis for each of the above principles in Chapter 12.
Researchers seeking to understand the causes of the rise in incarceration (discussed in detail in Chapters 3 and 4) have been able to identify and estimate the effects of several proximate causes, including specific state and federal policy choices. These include not only legislated policies but also changes in police practice, the behavior of prosecutors and judges, and the administration of parole, as well as other changes in how laws are implemented. These direct influences on the numbers incarcerated, however, have a social and historical context, including public concerns about crime and disorder, political incentives to respond to or exploit those concerns, and a complex history and evolution of racial and ethnic group relationships and politics.
Understanding the deeper sources of the rise in incarceration rates calls for other kinds of evidence and analysis than those applied in exploring proximate causes, along with a more subjective set of judgments about how to interpret that evidence. For example, it is not always possible to assess the motivation and incentives of leaders or voters that contributed to putting more people behind bars. The analytical problem is complicated by the largely autonomous decisions and actions taken by multiple actors in the states and the federal government. Nevertheless, the committee devoted considerable effort to exploring such questions of causality in the belief that a better understanding of how the United States reached this point is essential to understanding where the country should go from here.
Our analysis of both causes and consequences (discussed below) must remain provisional because the growth of incarceration is so recent, its effects are still unfolding, and the level and uses of incarceration at a given moment in time are the result of a complex set of past and ongoing social and political changes. Beginning in the 1960s, a complex combination of organized protests, urban riots, violent crime and drug use, the collapse of urban schools, and many other factors contributed to declining economic opportunities in many neighborhoods and too often to greater fear
of crime. After 1970, a wave of heavy industry closings and mass layoffs resulting from technological changes, international competition, and shifting markets contributed to the elimination of many relatively high-paying jobs for less educated workers with specialized skills. Places hardest hit by the wave of industrial losses also experienced large and sustained increases in crime (National Research Council, 2008, p. 5). In the 1980s, a wave of crack cocaine use and related street crime hit many of the nation’s already distressed inner cities.
In the wake of these and other structural shifts, employment fell among young people with little schooling and work experience. The income gap between unskilled and skilled workers widened. Large-scale migrations from rural to urban areas and an influx of lower-skilled undocumented workers from Mexico and Central America coincided with this widening inequality. In past decades, then, as growing percentages of all ethnic and racial groups have graduated from high school and went on to higher education, those who dropped out were left further behind with fewer options for earning a living.
In response to the protests and turmoil of the 1960s, major national commission reports examined the causes of racial division and the historical roots of violence in the United States (President’s Commission on Law Enforcement and the Administration of Justice, 1967; National Advisory Commission on Civil Disorders, 1968; U.S. National Commission on the Causes and Prevention of Violence, 1969; National Advisory Commission on Criminal Justice Standards and Goals, 1973). These and other experts recommended renewed efforts to address poverty and racial inequality at its roots. Following initial support for “Great Society” programs designed to tackle poverty, both an increase in violent crime starting in the 1960s and periodic civil disorder contributed to public fears of crime and support for tougher sanctions.
These fears and their political uses contributed in turn to a series of changes in criminal laws and prescribed punishments; changes in law enforcement and criminal justice procedures; and other changes affecting the frequency and severity of punishments, including incarceration. Pessimism emerged, among professionals as well as the larger public, about the potential of prisons to rehabilitate their occupants. Those affected most directly by changes in the criminal justice system lived in the neighborhoods most severely affected by the loss of economic opportunity and other kinds of distress, including high rates of violent crime and drug dealing. And in the past two decades of the period of the rise in incarceration rates, new attention to illegal immigration and to sex offenses led to new laws and penalties, contributing to growing numbers detained, convicted, and sentenced to prison for these offenses.
A full description of this complex and evolving social context for the expanded use of prison as a response to crime in the United States is beyond the scope of this study. Nonetheless, it is important to examine this history to better understand the rising use of imprisonment over nearly four decades. The committee does not view the growth of incarceration as an inevitable product of these larger social changes. The United States has experienced other periods of sweeping social change and disorder in which incarceration rates did not rise. During the recent period, moreover, the United States stood apart from other modern industrial democracies in the direction it took. To understand how U.S. incarceration rates reached their current level, the committee examined and weighed evidence of various kinds, drawing on the methods of historians, economists, and political scientists, as well as other social scientists.
The committee’s efforts to isolate the effects of the high rates of incarceration (discussed in Chapters 5 through 11) likewise faced a series of analytical challenges. For purposes of this study, we were interested in the aggregate effects of higher incarceration rates on individuals, their families and children, the communities from which they come, and U.S. society. Our ability to measure these effects depends in part on inferences drawn from numerous studies examining how the lives of those incarcerated differ from the lives of those otherwise like them who have not experienced incarceration. If, for instance, those who go to prison fare worse than others, is this then due to incarceration or some other factor, such as illiteracy or drug addiction, which is correlated with incarceration? Because the personal characteristics or behaviors that put people at greater risk of incarceration may cause them to do poorly in other ways, researchers may find it difficult to isolate the effects of incarceration.
To the extent that the effects of incarceration on individuals and their families can be measured, one may infer that a rise in incarceration rates will cause those effects to be more widespread. However, measurement of aggregate effects over time poses a second set of research challenges. One of these is the need to separate effects leading to or arising from the growth of incarceration in the United States from other, more or less contemporaneous social changes, such as changes in crime rates or labor market conditions. For example, if homicide rates fell in the United States after 1980, is it methodologically feasible to determine whether any part of that decline can be attributed to increased incarceration, taking into account all the simultaneous changes in other aspects of the society and the environment that have been shown to influence homicide? This was a central question
for the committee to address, but the methodological challenges make it a difficult one to answer.
As rates of many major crime types rose and fell for more than five decades, incarceration rates started to rise nationally in 1973 and continued to rise for 40 years, stabilizing only recently (Tonry, 2012; see Figure 1-1). Some may compare the decline in crime rates after 1980 and again in the 1990s with the increase in incarceration rates and infer that incarceration greatly reduces crime. However, studies of crime trends that consider many possible influences, including changes in incarceration rates, have had limited success separating different causes. Evidence concerning the complex relationship between crime and incarceration rates is reviewed in Chapter 5.
In a society or community where incarceration rates rise significantly, the experience of incarceration and/or its meaning for individuals and families may change, altering its effects. Given the demographic and geographic concentration of the rise in incarceration in the United States, the greatest impact has been felt by those living in the poorest communities. If the
FIGURE 1-1 U.S. crime and imprisonment rates, 1960-2010.
NOTE: The different crime rates have been rescaled, as noted in the figure, to facilitate comparison of time trends.
SOURCE: Tonry (2012).
incarceration experience has become more common in some communities than others, does this new reality alter the cumulative or marginal effects of incarceration on the rates of certain types of crime, or the deterrent effects of the criminal justice system, or the employment rates of those released back into those communities? As the rate of incarceration rises, its average and marginal effects may change as well, because those being locked up may include individuals who have different responses to incarceration relative to those imprisoned when the rate was lower. Therefore, an inquiry into the consequences of high rates of incarceration might focus on whether the incarceration of larger numbers has provided diminishing returns to U.S. society—smaller benefits at higher costs—at the margin. Finally, family dynamics, marriage, and labor markets may also function differently in communities with high rates of incarceration.
Another research challenge is posed by the diverse conditions of incarceration. The experience of incarceration varies from prison to prison, from state to state, and between the state and federal systems. It is also likely that the prison experience has changed over time, in particular over the period when incarceration rates rose and prison administrators were struggling to stay abreast of the rapid expansion of their systems. Furthermore, because incarceration is not a single, uniform experience but is highly varied, even to some degree unique to each prisoner, generalizing about its effects is difficult. Depending, among other things, on the conditions and duration of confinement and release, imprisonment can be humane and even helpful. Incarceration can provide time for reflection and personal growth, access to better health care and treatment of drug dependency or illness, respite from toxic environments or situations, and opportunities to learn and acquire skills. Alternatively, prison can be harmful and degrading, exposing the confined to criminal influence, violence, and humiliation; isolating them from nurturing human contact and personal responsibility; and leaving them poorly equipped for life outside. Generalization about prison’s effects requires more sophisticated measures of the nature of that experience than are often readily available to researchers, especially given ethical and practical barriers to conducting controlled studies inside prisons. Moreover, the prior characteristics of those incarcerated and their behavior help determine the nature of the prison experience both objectively and subjectively. Understanding the effects of incarceration thus requires specifying in some detail the nature of that experience, as well as distinguishing its effects from the effects of these preexisting characteristics and concurrent behaviors.
Not only is the base of knowledge about the prison experience constrained by a paucity of studies, but also even the extant literature may be of limited value. Because prison conditions and sentencing practices are continually evolving, sometimes in response to the level of incarceration and related budget pressures, some findings of earlier studies may not be
applicable to later periods. As typical prison conditions and the number and types of people exposed to them change over time, even strong empirical findings must be viewed as provisional. For example, if there has been a shift since the 1970s from a more rehabilitative to a more punitive prison management regime, the effects of incarceration may have changed as a result.
A further limit on analysis of prison conditions is imposed by barriers to researchers’ access to prisons. In earlier times, scholars were given broad access to prisons, prisoners, and corrections officers. By contrast, the contemporary prison has not been the subject of sustained empirical inquiry, perhaps reflecting less interest on the part of researchers and wariness on the part of prison administrators. The lack of research access, combined with cutbacks in other forms of outside review—journalistic investigations of life on the inside and judicial review of legal challenges to prison conditions—means that the nation’s prison systems are less open than before to public scrutiny. Finally, prisons have not been subjected to the same degree of regular reporting and public scrutiny based on transparent, published standards of performance as some other major public institutions, such as schools and hospitals.
Because the scientific study of a social phenomenon contributes to new understanding or interpretations, it has the potential to change the phenomenon itself. Indeed, a major purpose of the committee’s work is to provide the public and policy makers with new insights that could lead to new policies. In this period of evolving public opinion and policy and rapid expansion of the role of incarceration in U.S. society, there may be more reason than usual to regard all findings on the rise in the incarceration rate, including those of this study, as provisional and a new starting point for further research.
Summarizing research across a variety of disciplines—from history, to economics, to medicine, to law—and across a range of questions—from the historical sources of policy change to incarceration’s effects on individuals—defies any single standard of evidence or method of synthesis. Historians and economists, for example, work with different empirical materials, and each discipline treats social context differently. Assessing employers’ responses to formerly incarcerated job seekers, an area in which randomized trials are possible, enables strong causal inferences for which statistical certainty can be estimated. Analysis of the development of punitive crime policy, an area in which history is run just once in all its complexity, is no less rigorous and empirical, but the findings will be of a different nature. In executing a charge that ranged so widely, the committee
encountered methodological challenges in three areas: data, consequences, and historical interpretation.
Although the incarcerated population is reasonably well accounted for during our period of study, the data landscape often is seriously incomplete. There are no complete data series on the ethnicity and education of prison inmates, for example, so we frequently rely on estimates throughout this report. Longitudinal data following people in and out of prison and other contexts, such as schooling, work, and family, would have greatly improved our understanding of the effects of incarceration. Beyond these challenges, summarizing data from 51 prison jurisdictions (and thousands of localities) often involves glossing over the great institutional and statutory variation across the country. In this report, then, we attempt to paint a general national picture while recognizing that U.S. criminal justice is a matter largely for state and local governments. When we rely on case studies, we take care to extrapolate their lessons in light of the particular institutional context in which those lessons were generated. In general, many of the data gaps can be filled by using estimates or by pooling different data sources, but doing so necessarily adds uncertainty to the conclusions that are drawn. The major published data series on incarceration used in this report are described and assessed in Appendix B.
Besides providing an accurate portrait of the penal system, a key part of the committee’s charge concerned the consequences of incarceration. In a variety of different empirical domains—in the study of employment, health, families, and communities—incarceration is closely correlated with multiple measures of social and economic disadvantage. In rare cases, controlled experiments are possible. Just as rarely, so-called natural experiments—such as those arising from changes in law in some places but not others, or from variations in judicial interpretation or sentencing practice—produce variation in incarceration that is independent of the background conditions that are usually closely related to serving time in prison. More narrowly, the committee, like empirical researchers, was challenged to fully disentangle the effects of incarceration from those of conviction or other operational aspects of criminal justice, such as the conditions of confinement.
Finally, the committee confronted the problem of historical interpretation. For a large and regionally variable phenomenon such as the emergence of high incarceration rates, unfolding for more than four decades across 50 states and at the federal level, the historical record often does not point unambiguously to a single chain of events propelling its occurrence. As the historical narrative reaches back in time, from sentencing policy in the 1980s to the politics and social change of the 1960s and even earlier, the number of plausible rival explanations is multiplied. Both where causal inference about incarceration is threatened by correlated measures of socioeconomic disadvantage and where the historical record is open
to alternative interpretations, we have, whenever possible, tried to look at multiple sources of empirical evidence, generated by different research designs subject to different flaws and strengths. Pulling a variety of findings together, we have attempted to characterize the weight of the evidence, often describing the scientific consensus as a plausible range of conclusions instead of a unique inference.
Empirical studies have wrestled with the methodological challenges outlined above with varying success, so a major task of this committee was to separate stronger from weaker evidence. We have tried to describe carefully the limits of current knowledge so that readers of this report can form their own judgments about the strength of the evidence and therefore its usefulness as a basis for public policy and their own actions. We have taken care to indicate where uncertainty exists in research results, clarifying what is not known as much as what is.
Based on a weighing of the evidence and its limits, the committee’s final task was to assess its implications for public policy. As already noted, these implications depend on both normative and scientific analysis. People hold differing views regarding the outcomes of policy, including the balancing of risks to public safety against both the potential social benefits and reduced costs of alternatives to the policies that have led to the current high rate of incarceration.
Members of this committee share the view that the inherent severity of incarceration as a punishment—including the harm it often does to prisoners, their families, and communities—argues for limiting its use to cases where alternatives are less effective in achieving the same social ends. The evidence reviewed in this report reveals that the costs of today’s unprecedented rate of incarceration, particularly the long prison sentences imposed under recent sentencing laws, outweigh the observable benefits. We are conscious as well that a key feature of today’s high rate of incarceration is that large proportions of poor, less educated African American and Hispanic men are likely to be in jail or prison at some time in their lives. Indeed, for many poorly educated African American and Hispanic men, coercion is the most salient of their encounters with public authority. These findings led us to look for better policy choices.
The committee’s role is not to decide what level or use of incarceration is appropriate for the nation or a given state. Regardless of one’s values or preferences, however, choices can be informed by our assessment of the effects of and trade-offs among different policies. To the extent that trade-offs can be quantified, alternative policy regimes can be compared in terms of the levels of incarceration and public safety and other effects they may
produce. To the extent that evidence is conflicting or suggests a range of possible effects, alternative sets of assumptions will show a range of likely results of alternative policies.
The remainder of this report is organized around the major sets of questions posed in the committee’s charge. Chapter 2 provides an overview of the rise in the incarceration rate since the early 1970s, its components, and accompanying changes in society. Chapters 3 and 4 explore evidence regarding the complex set of factors contributing directly and indirectly to the high rate of incarceration in the United States. Chapters 5 through 11 examine evidence regarding the impacts of the rise in incarceration on crime (Chapter 5); the nature of the experience of prison for those incarcerated (Chapter 6); the effects of incarceration on prisoners’ health and mental health (Chapter 7) and employment and earnings after release (Chapter 8); the effects on children and families (Chapter 9) and on the communities from which prisoners come and to which they return (Chapter 10); and wider consequences for U.S. society (Chapter 11). Chapter 12 reviews the principles and values that determine the proper role of prisons in society and traces their intellectual lineage including the modern application of these foundational concepts. Chapter 13 summarizes the implications of the evidence on causes and consequences for public policy, applies the guidance provided by the principles that constrain the use of incarceration, and presents the committee’s findings and recommendations for policy makers and the public. This final chapter also presents our recommendations for research to address gaps in knowledge on many questions pertinent to the policy choices facing leaders and the public. A more detailed research agenda is provided in Appendix C.